Process polyT reads using ninetails pipeline

This function processes reads that have been classified as polyT-containing through the cDNA classification pipeline. Updated to work with tibble input instead of separate files. It applies the ninetails analysis pipeline to identify non-thymidine residues within polyT tails. Currently uses the same model as polyA processing until a polyT-specific model is trained.

Usage

process_polyt_reads_cdna(
  polyt_sequences,
  signal_files,
  num_cores = 1,
  qc = TRUE,
  save_dir,
  prefix = "",
  cli_log = message
)

Arguments

polyt_sequences

Data frame/tibble. PolyT-classified sequences with read_id and tail_type columns.

signal_files

Character vector. Paths to signal files containing poly(A) tail signals extracted from POD5 files (will be filtered for polyT reads).

num_cores

Integer. Number of CPU cores to use for parallel processing.

qc

Logical. Whether to apply quality control filtering.

save_dir

Character. Directory where processing results will be saved.

prefix

Character. Optional prefix for output file names.

cli_log

Function for logging messages and progress.

Value

List containing polyT processing results:

read_classes

Data frame with read classification results (with tail_type preserved)

nonadenosine_residues

Data frame with predicted modifications (with tail_type preserved)

processing_stats

Summary statistics for polyT processing

tail_type

Character indicating this is "polyT" data

Examples

if (FALSE) { # \dontrun{
polyt_results <- process_polyt_reads_cdna(
  polyt_sequences = polyt_tibble,
  signal_files = c("polya_signal_part1.rds"),
  num_cores = 4,
  qc = TRUE,
  save_dir = "path/to/output/",
  prefix = "experiment1",
  cli_log = message
)
} # }